Fialka: How the Soviet M-125 Cipher Machine Rebuilt the Rotor Idea

For most of the Cold War, a Western visitor would have had no way to see the machine that carried much of the Eastern Bloc’s secret military and diplomatic traffic. Its Soviet codename was Fialka (Фиалка, “violet”), its designation M-125, and its design was an answer to a question every postwar cryptographer was asking: if the German Enigma could be broken, what would a rotor machine have to look like to survive?

The answer the Soviet engineers produced kept Enigma’s basic architecture but changed nearly everything around it. This article sets out what the surviving machines and the specialists who have studied them can tell us, and marks clearly where the record runs out.

Origins and dates: what the sources say

The basic outline is not in dispute. Fialka was an electromechanical rotor cipher machine built in the Soviet Union after the Second World War and issued to the armed forces and governments of the Warsaw Pact.¹² The details of when it appeared vary from source to source, and the variation is worth stating plainly.

Crypto Museum, the Dutch collection run by Paul Reuvers and Marc Simons whose research underpins most of what is publicly known about the machine, says the M-125 was developed in the mid-1950s and introduced around 1956.¹³ A 2019 essay by Anna Borshchevskaya, published in Foreign Policy, also gives 1956 and says Fialka replaced an earlier Soviet machine called Albatross.⁴ A 2011 U.S. Air Force article describing an M-125 lent by the National Security Agency’s National Cryptologic Museum says the machine was “developed by the Soviet Union in the late 1950’s.”⁵ The Science Museum Group in London dates its own example to “c. 1958 or later,” which is a date for that object rather than for the design.²

The most defensible summary is that the M-125 was in service by the second half of the 1950s, with 1956 the date most often given. An improved model, the M-125-3, followed in the mid-1960s.⁶

The end date is similarly approximate. Borshchevskaya writes that Fialka remained in use “until the early 1990s,” while the Air Force article ties its use to the Soviet collapse in 1991.⁴⁵ No source located for this article gives documented retirement dates country by country.

Inside the machine: ten rotors and a punched card

At the heart of every Fialka is a drum of ten cipher rotors, more than double Enigma’s three or four. Each rotor has 30 contacts on each side, matching the 30 characters the machine handles; the M-125 was designed around the Russian language and the Cyrillic alphabet.³¹ Each of the ten rotors is wired differently.³

The rotors are built in two parts: an insert carrying the electrical contacts and scrambled wiring, and an outer ring fitted with mechanical pins. The presence or absence of a pin at a given position controls whether a neighboring wheel steps, which is why Crypto Museum calls them “advance blocking pins.”⁷³ The result is irregular stepping, quite unlike Enigma’s largely regular, odometer-like advance.

A second departure is that adjacent wheels turn in opposite directions. Seen from above, according to Crypto Museum, odd-numbered wheels move toward the front of the machine and even-numbered wheels toward the rear.³⁸

Before reaching the rotors, the electrical signal passes through a card reader on the left side of the machine. The reader holds a punched card laid out as a 30 by 30 matrix that rewires the 30 lines into a new permutation. Crypto Museum describes it as a substitute for Enigma’s plugboard, and notes that a new card was installed daily as part of the key.⁶⁹ Because protocol required the cards to be destroyed after use, very few originals survive.⁹

The machine was also a self-contained teleprinter terminal. It punched its output onto paper tape, printed it, and could read tape back in.³ On the original M-125, printing could not be switched off in punch mode, so tapes carried both holes and printed text.³ The M-125-3 added a 5-bit digital output at the rear for driving telegraph equipment or a transmitter directly, and a power supply that could run from a range of mains voltages or a truck’s 24-volt DC system.⁶

The fix for Enigma’s famous weakness

Fialka, like Enigma, uses a reflector: a fixed component at the end of the rotor stack that sends the current back through the rotors by a different path. In Fialka the reflector cannot be removed, and it connects the contacts in pairs, which makes the machine reciprocal (the same setting both encrypts and decrypts).¹⁰

On Enigma, that reflector had a well-known side effect. Because the current always returns by a path different from the one it entered on, a letter can never be encrypted as itself.¹¹ Polish and British codebreakers exploited that rule heavily, since it let them rule out possible placements of guessed plaintext.¹²

Fialka’s designers removed the guarantee deliberately. Crypto Museum describes what it calls the “magic circuit”: one wire pair in the reflector is broken, and one contact (number 13) is used to override the encoding so that a letter can come out as itself.¹¹ Isolating that contact leaves 29, an odd number that cannot all be paired.¹⁰ The machine therefore partly gives up perfect reciprocity in exchange for removing a pattern an attacker could use, and Crypto Museum puts the chance of any given letter enciphering to itself at 1 in 30.⁸¹⁰

So the frequently repeated summary “Fialka is a bigger Enigma” captures the family resemblance but misses the point. It shares the reflector and rotor concept, but adds:

• ten rotors instead of three or four;
• irregular, pin-controlled stepping;
• adjacent rotors turning in opposite directions;
• a daily punched-card permutation in place of a plugboard;
• a circuit that allows a letter to encrypt to itself.

One design, many national machines

Fialka was issued across the Warsaw Pact, but not as a single interchangeable machine. According to Crypto Museum, each country received its own keyboard, print head, and rotor wiring.¹ On the M-125-3 keyboards, Cyrillic characters sit in the same positions on every variant, while the Latin characters are arranged differently from country to country.¹³ The earlier M-125 could be switched between Russian and Latin operation.³

Crypto Museum’s list of M-125-3 variants is built on the Russian base model, the M-125-3M, and includes the M-125-3MR2 for Poland, M-125-3MN for East Germany, M-125-3MPx for Hungary, and M-125-3MR3 for Czechoslovakia.⁶ The Polish machine has a keyboard with Polish letters and its own print head and uses wheels with the prefix 3K; the Czechoslovak machine is electrically identical but has a different keyboard, print head, and 6K wheels.¹³ Crypto Museum reports finding rotor sets with five prefixes (68, 3K, 5K, 6K, and 7K) and says it has identified the countries for only some of them.¹ A quick way to tell the generations apart: the M-125 has rounded key tops, the M-125-3 square ones.³⁶

The national wiring had a practical consequence. Allied states could not read one another’s peacetime traffic on their own sets. Crypto Museum adds that if war with the West had become imminent, a common rotor set would have been distributed to all Warsaw Pact members.¹ That claim rests on Crypto Museum’s account; this article did not locate the underlying Soviet documentation.

Reports that Fialka was also used by communist states outside the Warsaw Pact, Cuba most often named, appear in secondary sources. Borshchevskaya refers more generally to “Warsaw Pact and other communist nations.”⁴ A specific list beyond the European Warsaw Pact members has not been confirmed from museum or archival records here.

From secret to museum piece

When the Soviet forces withdrew from Central and Eastern Europe, the machines were largely recovered and destroyed. Crypto Museum says only a small number escaped destruction and have since reached collectors and museums.¹

How the public record came to exist is a story of private research more than official release. Reuvers and Simons began studying Fialka in 2005, when, by their account, no technical information about it was available. Their Fialka Reference Manual, first issued in May 2005 and expanded to version 2.0 in 2009, runs to more than 140 pages covering the machine’s history, wheel wiring, and key cards, and has been free to download since 2021.¹⁴ Crypto Museum has continued to add material as machines surface. In late 2017 it reported examining an original Russian M-125-3 with its original Russian cipher wheels, a machine said to have been given to a senior KGB official on his retirement.¹⁵

Borshchevskaya writes that Russia declassified information about the machine in 2005.⁴ That date coincides with the start of the Crypto Museum project, and this article did not find an official Russian document recording a formal declassification. The safer statement is that detailed information became public from about 2005, largely through Western collectors’ work on surviving machines.

Today Fialkas can be seen in several public collections. The Science Museum Group holds a Cyrillic ten-rotor example with a spare rotor set in its metal drum.² The German Spy Museum in Berlin displays an M-125-3 with a Polish keyboard and presents it as the Soviet counterpart to Enigma.¹² The NSA’s National Cryptologic Museum lent an M-125 to the Air Force ISR Agency’s heritage center in San Antonio in 2011.⁵ In 2019, a specimen went on display at the KGB Espionage Museum in New York.⁴

What remains unknown

The single largest open question is whether any Western agency read Fialka traffic. No public source located for this article documents a NATO or NSA break of the machine, and none rules one out. Borshchevskaya reports the assessment of historian Stephen Budiansky, author of Code Warriors, that Soviet ciphers of this era could be defeated only through human error, theft, or defection rather than pure cryptanalysis.⁴ That is an informed judgment, not a published result of testing Fialka. Claims found online that the NSA captured a machine in 1967 and built a dedicated computer to attack it could not be traced to any identifiable source and should be treated as unsupported.

Other gaps are narrower. The exact service entry date, the full list of user states, the identity of every rotor-prefix series, and the operational procedures behind the daily key cards are known only in part, and much of what is known comes from one dedicated research team working from the hardware. That team’s work is careful and openly documented, but independent, peer-reviewed analysis of the machine’s security remains scarce. For a device that once guarded the communications of an entire military alliance, Fialka is still, in important respects, only half declassified.

Footnotes

1. Crypto Museum, “Fialka” overview page. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

2. Science Museum Group, object co8245733. ↩ ↩² ↩³

3. Crypto Museum, “Fialka M-125.” ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹

4. Borshchevskaya 2019, “The Soviets’ Unbreakable Code.” ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

5. U.S. Air Force 2011, “AFISRA debuts one-of-a-kind cryptologic display.” ↩ ↩² ↩³

6. Crypto Museum, “Fialka M-125-3.” ↩ ↩² ↩³ ↩⁴ ↩⁵

7. Crypto Museum, “Fialka cipher wheels.” ↩

8. Crypto Museum, “Fialka block diagram.” ↩ ↩²

9. Crypto Museum, “Fialka key card.” ↩ ↩²

10. Crypto Museum, “Fialka wiring.” ↩ ↩² ↩³

11. Crypto Museum, “Fialka: The Magic Circuit.” ↩ ↩²

12. Deutsches Spionagemuseum, Fialka M-125-3 object page. ↩ ↩²

13. Crypto Museum, Czechoslovak and Polish M-125-3 variant pages. ↩ ↩²

14. Reuvers and Simons 2009, Fialka Reference Manual 2.0, and its Crypto Museum description page. ↩

15. Crypto Museum, News 2017. ↩